Candida infections in neonates: Treatment and prevention

INTRODUCTION — Candida infections are common in neonates and have a broad range of clinical severity. In healthy term newborns, most candidal infections are mild (eg, oral thrush (picture 1), diaper dermatitis (picture 2A-B)). However, in preterm neonates, Candida is an important cause of invasive systemic illness and can be associated with substantial morbidity and mortality.

The treatment and prevention of Candida infections in neonates will be reviewed here. Related topics include:

●(See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis".)

●(See "Unusual fungal infections in the neonate".)

●(See "Candida infections in children".)

●(See "Common problems of breastfeeding and weaning", section on 'Candidal infection'.)

TREATMENT OF SUPERFICIAL MUCOCUTANEOUS INFECTIONS — Neonates with superficial mucocutaneous candidiasis (ie, oral thrush (picture 1), diaper dermatitis (picture 2A-B)) are generally at low risk of disseminated systemic infection and can be treated with topical antifungal therapy.

Appropriate topical antifungal agents for noninvasive mucocutaneous candidiasis include nystatin and azole topical preparations. Systemic antifungal agents are used only if the infection is refractory to topical therapy.

Oropharyngeal candidiasis (thrush)

●Who to treat – In healthy term infants who lack symptoms, are feeding well, and are at low risk of invasive candidiasis, oral candidiasis may not require treatment. Data concerning treatment in this setting are lacking, and the optimal approach is uncertain. Some experts prefer to treat all infants, including low-risk and asymptomatic neonates, whereas some guidelines suggest treating thrush in this setting only if there are symptoms (eg, feeding difficulties, maternal breast pain) [1]. Breastfed dyads can develop a cycle of breast infection in the mother, accompanied by oral thrush in the infant, with the infection repeatedly being transferred from one to the other. Simultaneous treatment of both may break this cycle. (See "Common problems of breastfeeding and weaning", section on 'Candidal infection'.)

●Choice of agent – Nystatin, a polyene topical antifungal agent, is our preferred initial agent to treat oral candidiasis (thrush) as it is not absorbed systemically from the gastrointestinal tract. It is administered as an oral suspension (100,000 units/mL) at a dose of 0.5 mL to each side of the mouth four times a day between feeds for 5 to 10 days. Reported cure rates with oral nystatin suspension range from 29 to 85 percent [2,3].

Oral miconazole gel (not available in the United States) has a better cure rate (>90 percent), but systemic absorption may occur. In one study involving preterm neonates, gastrointestinal side effects were reported in 6 percent [2].

For neonates with inadequate response to topical therapy (in combination with the decolonization/sterilization measures discussed below), we suggest treatment with oral fluconazole (3 to 6 mg/kg per dose once daily for 7 to 14 days).

●Prevention of reinfection – Measures to prevent reinfection include sterilization or decolonization of items that are placed in the infant's mouth (eg, bottle nipples and pacifiers) and treatment of Candida mastitis in the mother, if appropriate. (See "Common problems of breastfeeding and weaning", section on 'Candidal infection'.)

Diaper dermatitis — Candida diaper dermatitis is treated with topical nystatin or azole cream (eg, miconazole, ketoconazole, or clotrimazole cream) [2]. Nystatin usually is effective and is generally the least expensive option. Additional details are provided separately. (See "Diaper dermatitis", section on 'Candida superinfection'.)

TREATMENT OF INVASIVE SKIN INFECTIONS — Patients with invasive Candida skin infections (ie, congenital cutaneous candidiasis [CCC] (picture 3) or invasive fungal dermatitis) have a high likelihood of progressing to disseminated infection. For these patients, we suggest initiating systemic antifungal therapy while awaiting the results of blood and other cultures. A full evaluation should be performed to determine the extent of dissemination (table 1), which informs the duration of therapy. (See 'Evaluation for extent of disease' below.)

Congenital cutaneous candidiasis — Based on limited data, we suggest systemic antifungal therapy for neonates with CCC [4,5]. For premature infants and infants with complicated infection (clinical signs of sepsis or respiratory distress), amphotericin B is appropriate therapy [4]. Term infants who are tolerating feeds can be treated with oral fluconazole. (See 'Amphotericin B deoxycholate (conventional)' below and 'Fluconazole' below.)

Neonates with CCC are at risk for dissemination to the bloodstream, urine, or cerebrospinal fluid (CSF). CCC occurs most commonly in preterm low birth weight neonates, but it can also affect term newborns. The risk of dissemination is greatest in extremely low birth weight (ELBW; birth weight <1000 g) infants [4,5]. In a retrospective case series of 21 infants with CCC (18 preterm and 3 term), all but one infant received systemic antifungal therapy within 48 hours of rash presentation. Disseminated infection occurred in two infants (one was the infant in whom systemic antifungal therapy was delayed; the other presented shortly after birth with disseminated infection) [5]. There was one death in this series (the infant with early disseminated disease). These data provide limited evidence that promptly starting systemic therapy may reduce the risk of systemic infection and mortality. Earlier reports indicated that systemic infection developed in up to two-thirds of ELBW neonates with CCC, with mortality rates of up to 40 percent [4].

Invasive fungal dermatitis — Invasive fungal dermatitis is a condition that is unique to ELBW infants during the first two weeks after birth. Most affected patients have evidence of disseminated fungal disease. Thus, we suggest initiating systemic therapy with amphotericin B while awaiting culture results. (See 'Amphotericin B deoxycholate (conventional)' below and "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Invasive fungal dermatitis'.)

TREATMENT OF INVASIVE CANDIDIASIS

No role for routine empiric therapy — For neonates undergoing evaluation for suspected late-onset sepsis, it is generally not part of routine neonatal intensive care unit (NICU) practice to include an antifungal agent in the empiric antimicrobial regimen. While early empiric antifungal therapy (ie, prior to receiving culture results) theoretically might improve outcomes for neonates with candidemia, Candida accounts for only a small proportion of all late-onset bloodstream infections in this population. Thus, routine empiric coverage against Candida is generally not warranted. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Incidence' and "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Empiric antibiotic therapy'.)

However, selective use of prophylactic antifungal therapy may be warranted in high-risk patients (eg, extremely low birth weight [ELBW] neonates who are receiving a prolonged course of antibiotic therapy for confirmed bacterial sepsis), as discussed below. (See 'Targeted prophylaxis' below.)

In a multicenter observational study of 136 ELBW infants with invasive candidiasis, mortality rates were similar for infants who did or did not receive empiric antifungal therapy prior to culture results (33 versus 35 percent, respectively) [6]. However, those who received empiric antifungal therapy were more likely to survive without neurodevelopmental impairment (50 versus 36 percent).

Treatment of confirmed infection — For neonates with confirmed invasive Candida infection (ie, bloodstream infection, urinary tract infection [UTI], central nervous system [CNS] infection, and/or other site of invasive infection), management involves the following components [7-9]:

●Evaluation to assess the extent of dissemination. (See 'Evaluation for extent of disease' below.)

●Removal of any source of infection (eg, central venous catheter [CVC], urinary catheter).

●Administration of systemic antifungal agents. Antifungal therapy should be started as soon as possible once Candida has been identified in the blood culture or other specimen. In most cases, amphotericin B is the preferred drug for initial treatment. Once the results of susceptibility testing are available, the choice of antifungal agent can be modified, if appropriate. (See 'Amphotericin B deoxycholate (conventional)' below and 'Adjusting therapy based on susceptibility' below.)

Consultation with a pediatric infectious disease specialist is advised in most cases. Our suggested management approach outlined in the following sections is generally consistent with the recommendations of the Infectious Diseases Society of America and the American Academy of Pediatrics [7,9]. (See 'Society guideline links' below.)

Evaluation for extent of disease — Infants who are diagnosed with Candida infection involving the bloodstream, urinary tract, CNS, or other site of invasive infection should undergo a thorough evaluation to determine the extent of disease, which informs management decisions. The evaluation is summarized in the table (table 1) and discussed separately. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Evaluation for extent of disease'.)

Treatment based on extent of disease — The following sections outline our suggested treatment approach, including the choice of antifungal agent and duration of therapy, based upon the extent of disease. Additional details on specific antifungal agents, including dosing guidance and suggested monitoring, are summarized in the table (table 2), and discussed below. (See 'Systemic antifungal agents' below.)

●Candidemia without other organ involvement – For isolated bloodstream infection without evidence of other metastatic foci, we suggest treatment with intravenous (IV) amphotericin B for 14 days from the first negative culture [9]. (See 'Amphotericin B deoxycholate (conventional)' below.)

IV fluconazole is an acceptable option for susceptible isolates. (See 'Fluconazole' below and 'Adjusting therapy based on susceptibility' below.)

For infections that are associated with a CVC, management also includes removal of the CVC as soon as is feasible. The CVC serves as a focus for Candida, and removing it helps to clear the infection. In a retrospective study of 104 neonates with CVC-associated candidemia, early removal of the CVC (within three days of the first positive blood culture) was associated with shorter duration of candidemia (median of three versus six days) and lower mortality (0 versus 39 percent) compared with later or no CVC removal [10].

●Urinary tract infection (UTI) – Treatment of Candida UTI depends on whether there is evidence of kidney involvement and/or associated candidemia. All neonates diagnosed with Candida UTI warrant a complete evaluation to assess the extent of disease (including ultrasonography of the bladder and kidneys) (see 'Evaluation for extent of disease' above). In addition, for all neonates with Candida UTI, the urinary catheter, if present, should be removed. As with CVCs, the urinary catheter serves as a focus for Candida, and removing it helps to clear the infection.

•Simple cystitis (normal kidney ultrasound, no candidemia or other organ involvement) – For neonates with simple Candida UTI without kidney involvement (based on ultrasound), bloodstream infection, or other organ involvement, we suggest initial treatment with IV amphotericin B. (See 'Amphotericin B deoxycholate (conventional)' below.)

Once results of susceptibility testing are available, we suggest switching to IV or oral fluconazole if the isolate is susceptible. Oral therapy is appropriate if the neonate is clinically stable and tolerating feeds. (See 'Fluconazole' below and 'Adjusting therapy based on susceptibility' below.)

The usual duration of therapy for simple Candida cystitis is 10 to 14 days.

•UTI with candidemia – For neonates with Candida UTI associated with candidemia, treatment is the same as for those with isolated candidemia (ie, IV therapy for 14 days from the first negative culture).

•UTI with abscess or kidney involvement – If the ultrasound demonstrates kidney parenchymal infiltration, calyceal mycetoma, or fungal masses in the kidney or urinary tract, we typically treat with IV amphotericin B for at least 14 days and continue treatment until there is complete resolution by imaging. (See 'Amphotericin B deoxycholate (conventional)' below.)

Some studies suggest that ultrasonography correlates poorly with the course of the infection and patient outcome [11,12]. However, we continue to use ultrasonography to guide duration of therapy because alternative diagnostic/prognostic tools are not available. The clinical status of the patient is also an important consideration in determining the duration of treatment.

●CNS infection – For neonates with Candida CNS infection, we suggest IV amphotericin B for at least three weeks and until clinical signs, CSF abnormalities, and radiographic abnormalities (if present) have resolved [9]. (See 'Amphotericin B deoxycholate (conventional)' below.)

CNS devices (eg, shunts and ventriculostomy drains), if present, should be removed if possible [7].

If the CSF does not become sterile within a few days or if the patient's clinical status deteriorates despite treatment with amphotericin B monotherapy, we suggest adding flucytosine to the regimen [7,9]. (See 'Flucytosine' below.)

●Disseminated/multiorgan infection – Disseminated candidemia associated with other organ involvement (eg, endocarditis, fungal abscesses) is often difficult to eradicate. Prolonged antifungal therapy is generally warranted in these infants (usually four to six weeks). Surgical removal of the foci may also be necessary in some cases [7]. The duration of therapy is individualized based upon sterilization of blood cultures, resolution of clinical signs, and resolution of imaging findings. The decision should be made in consultation with a pediatric infectious disease specialist.  

Amphotericin B is generally the preferred agent for initial therapy of disseminated candidiasis, though the choice ideally should be guided by the susceptibility pattern of the isolate. Fluconazole is an acceptable option if the isolate is susceptible. (See 'Amphotericin B deoxycholate (conventional)' below and 'Fluconazole' below and 'Adjusting therapy based on susceptibility' below.)

Adjusting therapy based on susceptibility — In all patients with positive cultures, the Candida species and its susceptibility to antifungal agents should be determined. Based upon the results, the choice of antifungal agent can be modified, if appropriate. If the species is susceptible to amphotericin B, monotherapy is the preferred choice of therapy. Fluconazole is an alternative agent for susceptible isolates. (See 'Amphotericin B' below and 'Fluconazole' below.)

Amphotericin B is effective for treating most neonatal invasive Candida infections. However, some Candida species are less susceptible (eg, Candida glabrata and Candida krusei) or resistant (eg, Candida lusitaniae) to amphotericin B. C. auris is often multidrug resistant. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Microbiology'.)

Persistent infection — For patients with persistent candidemia that does not clear despite treatment with a single antifungal agent, we suggest adding a second agent. For patients initially treated with amphotericin B, options for add-on therapy include fluconazole or an echinocandin (eg, caspofungin, micafungin). There is a paucity of evidence to guide the best combination therapy in such neonatal cases [13]. (See 'Fluconazole' below and 'Echinocandins' below.)

Additional steps to take in patients with persistent or refractory invasive candidiasis include:

●All potentially contaminated indwelling foreign material should be removed (eg, CVCs, urinary catheters, ventriculostomy drains).

●Follow-up imaging studies (eg, ultrasonography of the head, liver, spleen, kidneys, and bladder) should be obtained to reassess findings seen on the initial evaluation and/or identify new focal sources of infection. Other imaging modalities (eg, computed tomography, magnetic resonance imaging) may be required in some cases.

●Surgical resection of infected tissue may be needed if systemic antifungal therapy is not successful in eradicating the infection or if the infected tissue causes functional impairment (eg, a urinary fungal mass causing urinary obstruction or a right atrial mass causing hemodynamic compromise) [14-17].

PREVENTION OF CANDIDA INFECTIONS IN PRETERM NEONATES — For preterm neonates, prevention of systemic Candida infections is important because candidemia is associated with considerable morbidity and mortality in this population (see 'Outcome' below). Prevention efforts have focused on extremely low birth weight (ELBW; birth weight <1000 g) infants who are at the highest risk for invasive Candida infections. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Risk factors for invasive candidiasis'.)

General measures — General measures are focused on decreasing cross-infection of Candida within neonatal intensive care units (NICUs) and reducing risk factors that increase the likelihood of candidemia. These measures include:

●Hand hygiene and gloves – Handwashing, use of gloves, and avoidance of artificial fingernails can reduce the rate of horizontal transmission from health care workers to infant. However, these measures alone may not be sufficient to prevent horizontal transmission, because of the difficulty in eradicating Candida from the hands of health care workers. In one study, commonly used antiseptics or disinfectants for handwashing demonstrated varying degrees of in vitro inhibition of growth of Candida species recovered from hospitalized patients [18]. (See "Infection prevention: Precautions for preventing transmission of infection", section on 'Hand hygiene'.)

●Cohorting and isolation – For infants who are known to be colonized or infected with Candida, cohorting or using single-room isolation may reduce the risk of colonization and infection in other neonates within the same NICU. With this intervention, health care workers can be segregated to only caring for infants with candidal colonization/infection and caring for noninfected/noncolonized infants. Although patient isolation or cohorting is often a recommended routine infection control measure, there are no clinical trials demonstrating that this practice prevents Candida cross-infection in the NICU [19].

●Minimizing or avoiding interventions that increase the risk of invasive candidiasis – Minimizing or avoiding important risk factors for candidiasis (eg, broad spectrum antibiotics, invasive devices, glucocorticoids, gastric acid suppressing drugs), if clinically appropriate, may decrease the risk of candidemia. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Risk factors for invasive candidiasis'.)

Removing central venous catheters once they are no longer needed helps to reduce all types of nosocomial bloodstream infections, including candidemia. Nutrition policies that promote early initiation and advancement of enteral feeding may reduce central line days. (See "Parenteral nutrition in premature infants" and "Approach to enteral nutrition in the premature infant".).

Targeted prophylaxis — Based upon the available evidence, our approach is to use a targeted antifungal prophylaxis rather than routinely providing prophylaxis in all preterm infants. This approach is consistent with the recommendations of the American Academy of Pediatrics and the Infectious Diseases Society of America [7,9,20]. Broad use of antifungal prophylaxis for all preterm infants is discouraged because of the concern of promoting resistant Candida species within the NICU. Universal prophylaxis in a NICU with a low incidence of candidiasis would expose a large number of infants to fluconazole to prevent a single case of infection and hence not recommended.

●Patient selection – Targeted antifungal prophylaxis is reserved for ELBW infants who are at highest risk of invasive Candida infection. This includes:

•ELBW infants cared for in NICUs with a high baseline rate of systemic fungal infection (ie, >10 percent), or

•ELBW infants who require antibiotic therapy for >48 hours (eg, for a confirmed bacterial infection or necrotizing enterocolitis)

●Choice of agent — When antifungal prophylaxis is indicated, we suggest fluconazole rather than other agents. (See 'Fluconazole' below.)

We prefer fluconazole over other agents because it is well studied and has demonstrated efficacy in this setting [21]. Oral or topical nystatin or miconazole are reasonable alternatives.

●Dosing and timing of administration – The dose for fluconazole prophylaxis is 6 mg/kg per dose given intravenously (IV) or enterally twice a week [9,22,23] The timing of administration depends on the indication. For ELBW neonates receiving prophylaxis due to high rates of invasive Candida infections within the NICU, fluconazole is started within the first 48 to 72 hours after birth and continued for four to six weeks or until the infant no longer requires IV access. For those receiving prophylaxis while on broad-spectrum antibiotics, fluconazole is administered throughout the antibiotic course.

●Supporting evidence – The practice of using antifungal prophylaxis in high-risk preterm neonates is supported by clinical trials and meta-analyses [21,24,25]. In a meta-analysis of 10 trials including 1371 preterm infants, systemic antifungal prophylaxis (all but one trial used fluconazole) reduced the incidence of invasive fungal infection compared with placebo or no drug (6 versus 16 percent; relative risk [RR] 0.43, 95% CI 0.31-0.59); the effect on mortality was not statistically significant (13 versus 17 percent; RR 0.79, 95% CI 0.61-1.02) [21]. The trials in the meta-analysis were generally of good methodologic quality; however, the incidence of baseline invasive fungal infection in these trials (16 percent) was much higher than that reported from large cohort studies (<5 percent). (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Incidence'.)

Studies directly comparing fluconazole with nystatin are limited. In a meta-analysis of three small trials, the overall rate of invasive fungal infection was lower with fluconazole compared with nystatin; however, the difference was not statistically significant (3 versus 6 percent; risk ratio 0.53, 95% CI 0.19-1.52) [21].

Several trials have shown that prophylactic use of oral or topical nonabsorbed agents reduces the risk of candidiasis [26-30]. In a meta-analysis of four trials including 1800 infants, prophylactic therapy reduced the incidence of invasive fungal infection compared with placebo (RR 0.2, 95% CI 0.14-0.27); the effect on mortality was not significant (RR 0.87, 95% CI 0.72-1.05) [26]. However, the trials in the meta-analysis had important methodologic limitations (including quasi-randomization and lack of blinding), which decreases the certainty of these findings.

Strategies not recommended — Other strategies that have been investigated for prevention of Candida infections in preterm neonates include performing routine surveillance cultures, using supplementation with immunonutrients (eg, lactoferrin), and probiotics. Based on the available data, we suggest not using these strategies.

●Surveillance cultures – Systematic screening for Candida colonization has been suggested as a possible tool in preventing fungal infections in high-risk patients by initiating antifungal therapy when colonization is detected [31]. However, this approach is not routinely used in most NICUs since colonization is very common, whereas invasive infection occurs in only a small percentage of colonized infants. Thus, the clinical utility of documenting Candida colonization is uncertain. (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Colonization'.)

●Lactoferrin – Based on the available evidence, it remains uncertain if lactoferrin supplementation meaningfully reduces invasive fungal infection in preterm infants. If there is an effect, it appears to be small. In a 2020 meta-analysis of six trials (3266 infants), enteral lactoferrin supplementation decreased invasive fungal infection compared with placebo (0.4 versus 1.7 percent; RR 0.23, 95% CI 0.10-0.54) [32]. However, there were few events (33 total episodes in the six trials), the absolute effect size was small (absolute risk difference 1.3 percent [98% CI 0.8-1.5]), and two of the trials had important methodologic limitations that may have biased the findings. In the largest trial, which involved >2200 preterm neonates randomized to enteral bovine lactoferrin or placebo (sucrose), rates of documented invasive fungal infection were similar in both groups (0.3 versus 0.2 percent, respectively) [33].

●Probiotics – The use of probiotics to prevent fungal and bacterial infections in neonates has been studied in randomized clinical trials [34]. However, important uncertainties and safety concerns remain, and we suggest not routinely using probiotics for this purpose.

In a meta-analysis of two trials (329 patients), probiotics reduced rates of Candida colonization in preterm infants (12 versus 31 percent; RR 0.43, 95% CI 0.27-0.68) [34]. Despite the apparent large effect in reducing colonization, probiotics did not appear to have a meaningful impact on reducing the incidence of invasive fungal infection (1.8 versus 2 percent; RR 0.88, 95% CI 0.44-1.78; six trials, 1259 patients). The studies included in the meta-analysis varied considerably with regard to the probiotic product used and birth weight and gestational age of enrolled infants. In addition, the individual trials were relatively small (80 to 400 patients) and many were assessed as having high risk of bias.

Other important uncertainties remain, including optimal probiotic strains, doses, and duration of therapy. Appropriate regulatory control of these products is another unresolved issue. In addition, rare but serious cases of probiotic-associated sepsis have been reported [35-38]. Until it has been shown that this intervention is beneficial and safe, and an effective standardized regimen has been developed, the use of probiotics for preventing fungal infections cannot be recommended.

Studies investigating the use of lactoferrin and probiotics for the prevention of bacterial sepsis in preterm neonates are discussed in greater detail separately. (See "Treatment and prevention of bacterial sepsis in preterm infants <34 weeks gestation", section on 'Potential prophylactic therapy'.)

SYSTEMIC ANTIFUNGAL AGENTS — The four major classes of antifungal agents used to treat invasive candidiasis in neonates are (table 2) [39]:

●Polyenes – Polyenes include amphotericin B deoxycholate, which is the preferred antifungal agent for most invasive neonatal Candida infections [7,9]. (See 'Amphotericin B deoxycholate (conventional)' below.)

●Triazoles – The most commonly used triazole in neonates is fluconazole. It is an alternative to amphotericin B if the isolate is susceptible. (See 'Triazoles' below.)

●Nucleoside analogues – The most commonly used nucleoside analogue is flucytosine. It is sometimes used in combination with amphotericin B in neonates with CNS infections. (See 'Flucytosine' below.)

●Echinocandins – Echinocandins include micafungin, caspofungin, and anidulafungin. These drugs are not routinely used for neonatal Candida infections. (See 'Echinocandins' below.)

Amphotericin B

Amphotericin B deoxycholate (conventional) — Amphotericin B deoxycholate is the preferred drug for treatment of most systemic neonatal Candida infections [9]. Most Candida species are susceptible to amphotericin B, and this drug is generally well tolerated in neonates [9,40-42].

Dosing and administration — Amphotericin B is administered intravenously (IV) at a dose of 1 mg/kg every 24 hours [9,41].

The dose of amphotericin B generally does not need be adjusted in patients with preexisting kidney impairment. However, in the unlikely event that the neonate develops drug-related acute kidney injury (AKI) while on therapy, the dose should be reduced (either by using alternate-day dosing or reducing the daily dose by 50 percent).

Adverse effects — Adverse effects of amphotericin appear to be less common and less severe in neonates than in older children and adults. Although there is potential for nephrotoxicity and hypokalemia due to kidney losses, it appears that most infants display no or only transient nephrotoxicity [43]. This was illustrated in a retrospective review of 92 infants <90 days of age (median gestational age 26 weeks) who received at least three doses of amphotericin B [43]. AKI developed in 16 percent of patients. AKI was mild (defined as serum creatinine 0.4 to 1.5 mg/dL) in 13 percent and more severe (serum creatinine >1.5 mg/dL) in 3 percent. On-treatment creatinine values did not exceed 2 mg/dL in any patients. However, one patient required discontinuation of amphotericin due to worsening AKI and oliguria (the patient was also receiving prolonged aminoglycoside therapy, which likely contributed to the AKI). In addition, 17 percent of patients developed hypokalemia (defined as serum potassium <3 mEq/L), which was more common in patients with AKI. By the end of amphotericin therapy, creatinine levels returned to baseline in eight patients, were decreasing in three patients, and remained elevated in four patients.

Other reported adverse effects include hypomagnesemia caused by excessive kidney losses, bone marrow suppression with anemia and thrombocytopenia, and elevated hepatic enzymes. These abnormalities are infrequent, dose dependent, and resolve with stopping the drug.

Monitoring — Because of the potential adverse effects, infants receiving amphotericin B should have regular laboratory monitoring. In our practice, we monitor the following laboratory studies:

●Serum electrolytes (including potassium and magnesium) daily or every other day initially, then once or twice weekly if stable

●Serum creatinine daily or every other day initially and, if stable, testing can be spaced to once or twice weekly

●Complete blood count with differential and platelet count twice a week initially and, if stable, testing can be spaced to once a week

●Liver enzymes (alanine aminotransferase [ALT] and aspartate aminotransferase [AST]) once after starting treatment and, if normal, ongoing testing is performed once a week

Other amphotericin B formulations — We suggest conventional amphotericin B (ie, amphotericin B deoxycholate) over the lipid complex or liposomal amphotericin B formulations. These formulations do not appear to be superior to the less costly amphotericin B deoxycholate. They are typically reserved for patients who develop intolerant infusion-related reactions or AKI during conventional amphotericin B administration. Lipid complex and liposomal formulations should not be used in patients with Candida kidney infections since these agents do not achieve adequate penetration into the kidneys [44].

Lipid complex and liposomal amphotericin B formulations have the ability to deliver a higher dose of medication with lower risk of toxicity but they are more costly than conventional amphotericin B. Data on the use of these agents in neonates are limited to case reports and case series [45-49]. Most reports show that these amphotericin B preparations successfully treated neonates with candidemia and were not associated with any major adverse effects. However, in a multicenter retrospective review of 730 infants with positive blood, urine, or cerebrospinal fluid (CSF) cultures for Candida species, multivariate analysis showed higher mortality for infants receiving lipid complex or liposomal amphotericin B lipid formulations compared with those receiving amphotericin B deoxycholate (odds ratio [OR] 1.96, 95% CI 1.16-3.33) or fluconazole (OR 2.39, 95% CI 1.18-4.83) [44].

Triazoles

Fluconazole — Fluconazole, a first-generation triazole, is an alternative to amphotericin B in neonates with susceptible isolates [9]. Fluconazole has several advantages:

●Excellent agent for the treatment of Candida UTI because it is excreted unchanged in high concentrations into the urine

●Excellent bioavailability when taken orally, potentially reducing the need for IV administration

The major disadvantage of this agent is the emergence of fluconazole-resistant Candida species, such as C. krusei, C. glabrata, and Candida parapsilosis [50]. The use of fluconazole prophylaxis to reduce Candida infections may also be a contributing factor to fluconazole resistance. (See 'Targeted prophylaxis' above.)

Thus, fluconazole should not be used as the agent for initial treatment of an infant with candidemia or suspected candidemia before the isolate's susceptibilities are known. It should be given only after identifying the Candida species and determining that the organism is susceptible to fluconazole. This is especially important if the neonate previously received fluconazole prophylaxis.

Dosing and administration

●Treatment dosing – For treatment of systemic candidiasis, fluconazole dosing for susceptible isolates consists of an initial loading dose of 25 mg/kg IV on day one, followed by 12 mg/kg per dose IV once daily [9,23]. IV administration is generally preferred over oral for the treatment of invasive infections. Oral fluconazole may be appropriate for treatment of isolated uncomplicated UTI and for mucocutaneous candidiasis, including congenital cutaneous candidiasis (CCC) in term infants who are clinically stable and tolerating full enteral feeding. (See 'Treatment based on extent of disease' above.)

●Prophylactic dosing – When fluconazole is used for prophylaxis, it is given at a dose of 6 mg/kg IV or orally twice weekly. Additional details regarding prophylaxis are provided above. (See 'Targeted prophylaxis' above.)

Side effects and monitoring — Fluconazole is generally well tolerated in neonates and side effects are uncommon. The main potential adverse effect is a transient elevation in liver enzymes, which generally resolves after discontinuing therapy. We suggest monitoring liver function tests twice weekly initially, and, if stable, testing can be spaced to once a week. No monitoring is necessary for neonates receiving fluconazole at prophylactic doses.

As previously discussed, another adverse effect of this agent is the potential for emergence of fluconazole-resistant Candida species.

Efficacy — The efficacy of fluconazole for treatment of invasive candidiasis is supported by two small prospective studies [51,52]. In a randomized trial of 23 infants with proven fungal septicemia randomly assigned fluconazole (IV or enterally) or amphotericin B, infants in the fluconazole group required fewer days of IV therapy [52]. However, mortality rates were high in both groups (3 of 12 patients in the fluconazole group versus 5 of 11 patients in the amphotericin B group). Patients who received fluconazole appeared to have less hepatotoxicity with lower serum bilirubin and alkaline phosphatase concentrations [23,53].

Studies evaluating the efficacy of prophylactic fluconazole are discussed above. (See 'Targeted prophylaxis' above.)

Second generation triazoles — Second-generation triazoles (eg, voriconazole, posaconazole) have been developed with a broader spectrum of activity and increased potency. There are few data on these agents in neonates. Their use in children is discussed separately. (See "Candidemia and invasive candidiasis in children: Management", section on 'Azole antifungal agents'.)

Flucytosine — In neonates, use of flucytosine is mostly limited to treatment of CNS infections that do not respond to monotherapy with amphotericin B. In this setting, flucytosine is used in combination with amphotericin B because it has excellent penetration into the CSF and is synergistic with amphotericin B [40].

Drawbacks of this agent in neonates include its considerable side effect profile, rapid development of resistance when used as monotherapy, and lack of availability of a parenteral formulation, which may limit its use in critically ill neonates. (See "Pharmacology of flucytosine (5-FC)", section on 'Adverse effects'.)

The usual dose is 25 mg/kg per dose given orally every six hours. It is critical to monitor blood levels of flucytosine to avoid bone marrow toxicity. Peak plasma concentrations obtained 60 to 120 minutes after administration should be maintained between 40 and 60 mcg/mL, and trough levels obtained prior to the next dose should be less than 25 mcg/mL. Bone marrow suppression is increased with sustained levels greater than 100 mcg/mL.

Echinocandins — Echinocandins (eg, micafungin, caspofungin, anidulafungin) are not first-line antifungal agents for treatment of neonatal invasive candidiasis since data on the safety and efficacy of these agents in neonates is very limited. These agents are reasonable options for treating Candida infections when neither amphotericin nor fluconazole can be used (eg, due to antifungal resistance and/or drug toxicity) [54]. In particular, C. auris is often multidrug resistant, and echinocandins are often used in combination with other agents to treat these infections. Micafungin and caspofungin are the agents that are most commonly used in these cases.

Resistance to this class of antifungal agents is uncommon. However, echinocandins should not be used for treating infections involving the eye, central nervous system, or urinary tract in the neonate since they have relatively poor penetration at these sites.

While echinocandins appear to be a promising class of antifungal agents in the treatment of neonatal candidemia, further data on their efficacy, safety, and dosing regimen in neonates are needed before echinocandins can be recommended for routine use.

●Micafungin – The efficacy and safety of micafungin in neonates are supported by observational data and small clinical trials [55-57]. In a systematic review of nine clinical trials evaluating the safety and efficacy of micafungin in infants <2 years (116 patients, 48 percent preterm infants), micafungin had good efficacy (73 percent treatment success among patients with invasive candidiasis) and was well tolerated without serious side effects [56]. In a subsequent clinical trial in 30 infants (most were ≤4 weeks old) with invasive candidiasis randomized to treatment with micafungin or amphotericin B, the rate of complete or partial clinical response after one week was similar in both groups (61 and 70 percent, respectively) [57]. The rate of treatment-associated adverse events was also similar in both groups (55 and 50 percent, respectively), with transient anemia and thrombocytopenia being the most common.

●Caspofungin – The efficacy and safety of caspofungin in neonates are supported by observational data and small clinical trials [58-61]. A small multicenter randomized trial compared caspofungin with amphotericin B in 47 neonates and young infants with invasive Candida infection [61]. At two weeks, the proportion of infants who were alive and free of infection was similar in both groups (71 versus 69 percent); adverse events occurred less commonly in the caspofungin group (85 versus 100 percent). However, the trial was stopped early due to low enrollment and, as such, it may have been inadequately powered to detect important differences in outcomes.

OUTCOME — The prognosis for neonates with invasive Candida infections depends on the severity of the infection, the gestational age, and other comorbidities. Among extremely low birth weight (ELBW; birth weight <1000 g) infants, reported mortality rates range from 20 to 40 percent [62-70]. Survivors of neonatal candidemia, especially in those with central nervous system (CNS) involvement, are at risk of long-term neurodevelopmental impairment [62-65,71-73]. (See "Long-term neurodevelopmental impairment in infants born preterm: Epidemiology and risk factors".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Candidiasis".)

SUMMARY AND RECOMMENDATIONS

●Superficial mucocutaneous candidiasis (oral thrush, diaper dermatitis) – Superficial mucocutaneous Candida infections can be treated with topical antifungal agents (see 'Treatment of superficial mucocutaneous infections' above):

•For treatment of oropharyngeal candidiasis (oral thrush), we suggest treatment with oral nystatin rather than other agents (Grade 2C). Nystatin is administered as an oral suspension (100,000 units/mL) at a dose of 0.5 mL to each side of the mouth four times a day between feeds for 5 to 10 days. Oral miconazole gel (not available in the United States) is an alternative option, but it has a high rate of side effects in preterm infants. No specific treatment is a reasonable alternative in healthy term infants who lack symptoms, are feeding well, and are at low risk of invasive candidiasis. (See 'Oropharyngeal candidiasis (thrush)' above.)

•Treatment of Candida diaper dermatitis is discussed separately. (See "Diaper dermatitis", section on 'Candida superinfection'.)

●Invasive skin infections – Patients with invasive Candida skin infections (ie, congenital cutaneous candidiasis (picture 3) or invasive fungal dermatitis) have a high likelihood of progressing to disseminated infection. For these patients, we suggest initiating systemic therapy with amphotericin B while awaiting the results of blood and other cultures (Grade 2C). (See 'Treatment of invasive skin infections' above and 'Evaluation for extent of disease' above and 'Amphotericin B deoxycholate (conventional)' above.)

●Invasive candidiasis – Neonates with invasive Candida infections should undergo thorough evaluation to assess the extent of disease (table 1). (See "Candida infections in neonates: Epidemiology, clinical manifestations, and diagnosis", section on 'Evaluation for extent of disease'.)

The choice and duration of therapy depends on the extent of disease. (See 'Treatment based on extent of disease' above and 'Systemic antifungal agents' above.)

•Choice of agent – For most neonates with invasive Candida infection, we suggest amphotericin B for initial therapy rather than other antifungal agents (Grade 2C). (See 'Amphotericin B deoxycholate (conventional)' above.)

Once results of susceptibility testing are available, therapy can be modified, if appropriate. IV fluconazole is an acceptable option if the isolate is susceptible. IV therapy is preferred over oral therapy for most invasive infections with the exception of simple cystitis, which can be treated with oral fluconazole if the neonate is clinically stable and tolerating feeds. (See 'Adjusting therapy based on susceptibility' above and 'Fluconazole' above.)

Guidance on dosing and monitoring for these agents is provided in the table (table 2).

•Duration of therapy – The duration of therapy depends on the type of infection (see 'Treatment based on extent of disease' above):

-Uncomplicated candidemia – – For neonates with candidemia (including CVC-associated candidemia) who lack evidence of metastatic foci of infection, IV therapy is provided for 14 days from the first negative blood culture. If a CVC is present, it should be removed as soon as possible.

-UTI – For neonates with simple Candida cystitis (ie, UTI without associated candidemia, evidence of kidney involvement, or other metastatic foci), the usual treatment duration is 10 to 14 days. If the ultrasound shows kidney involvement or a fungal mass, IV therapy is continued for at least 14 days or until there is complete resolution by imaging.

-CNS infection – Treatment is provided for a minimum of three weeks and should be continued until all clinical signs, cerebrospinal fluid (CSF) abnormalities, and radiographic abnormalities (if present) have resolved. CNS devices (eg, shunts and ventriculostomy drains), if present, should be removed as soon as is feasible. If the CSF does not become sterile within a few days or if the patient's clinical status deteriorates despite monotherapy with amphotericin B, we suggest adding flucytosine (Grade 2C). (See 'Flucytosine' above.)

-Disseminated disease/multiorgan involvement – Candidemia associated with metastatic foci of infection (eg, endocarditis, fungal abscesses) is often difficult to eradicate. The duration of therapy is individualized based upon sterilization of blood cultures, resolution of clinical signs, and resolution imaging findings. Prolonged therapy is generally warranted. Surgical removal of infected foci may be necessary in some cases. (See 'Persistent infection' above.)

●Prevention – In preterm neonates, strategies to prevent systemic Candida infections include (see 'Prevention of Candida infections in preterm neonates' above):

•General measures – General preventive measures include efforts aimed at reducing Candida cross-infection (eg, handwashing, use of gloves, isolation, cohorting) and efforts aimed at reducing the risk factors for candidemia (eg, antibiotic stewardship limiting use of broad-spectrum antibiotics, removing CVCs once they are no longer needed). (See 'General measures' above.)

•Selective use of antifungal prophylaxis in high-risk neonates – For extremely low birth weight (ELBW; <1000 g) neonates who are at high risk of Candida infection (ie, those receiving antibiotic therapy for >48 hours and/or cared for in a center with a high baseline rate of candidiasis [≥10 percent]), we suggest prophylactic fluconazole (Grade 2B). (See 'Targeted prophylaxis' above.)

●Outcome – The prognosis for neonates with invasive Candida infections depends on the severity of the infection, the gestational age, and other comorbidities. Among ELBW infants, reported mortality rates range from 20 to 40 percent. Survivors of neonatal candidemia, especially in those with CNS involvement, are at risk of long-term neurodevelopmental impairment. (See 'Outcome' above.)